Package Details: libvpx1.3 1.3.0-4

Git Clone URL: https://aur.archlinux.org/libvpx1.3.git (read-only, click to copy)
Package Base: libvpx1.3
Description: VP8 and VP9 codec
Upstream URL: http://www.webmproject.org/
Licenses: BSD
Provides: libvpx.so
Submitter: City-busz
Maintainer: patlefort
Last Packager: patlefort
Votes: 6
Popularity: 2.28
First Submitted: 2025-10-28 10:53 (UTC)
Last Updated: 2025-10-29 12:27 (UTC)

Dependencies (3)

Required by (17)

Sources (2)

Latest Comments

Riedler commented on 2025-11-27 13:57 (UTC)

ah, thanks for the info

patlefort commented on 2025-11-27 12:56 (UTC)

You aren't supposed to use -flto in your C(XX)FLAGS. Use LTOFLAGS instead.

Riedler commented on 2025-11-27 12:49 (UTC)

doesn't build with -flto in CFLAGS/CXXFLAGS (unsure why that isn't filtered by '!lto'…) and extra doesn't build with clang. please blacklist those if possible

berton commented on 2025-11-11 19:26 (UTC)

It built from the clean chroot so installed it on my host using the chroot-produced files (with makepkg -si), however I have no idea what was wrong...

patlefort commented on 2025-11-11 18:20 (UTC)

Try building in a clean container: https://wiki.archlinux.org/title/DeveloperWiki:Building_in_a_clean_chroot

berton commented on 2025-11-11 17:28 (UTC) (edited on 2025-11-11 17:28 (UTC) by berton)

I'm getting this error when building:

make[1]: *** No rule to make target 'vp8/encoder/vp8_asm_enc_offsets.c.S', needed by 'vp8_asm_enc_offsets.asm'.  Stop.

noxie commented on 2025-10-31 11:10 (UTC)

@patlefort hey!

After digging, the issue arises from pr when using uutils coreutils. Using gnu ones it works.

Wasn't hardware nor clean build (which it already was).

I'll raise the issue directly with uutils.

patlefort commented on 2025-10-31 10:52 (UTC)

I'm unable to reproduce on any of my machines. Did you try in a clean container? (https://wiki.archlinux.org/title/DeveloperWiki:Building_in_a_clean_chroot)

Also, what hardware?

noxie commented on 2025-10-31 08:51 (UTC)

As @smn commented yesterday I can confirm, it fills the whole memory (>64GiB) and the swap, and then the kernel kills the process.

==> Starting build()...
Configuring selected codecs
  enabling vp8_encoder
  enabling vp8_decoder
  enabling vp9_encoder
  enabling vp9_decoder
Configuring for target 'x86_64-linux-gcc'
  enabling x86_64
  enabling pic

this is exactly where it starts filling everything up.

smn commented on 2025-10-30 14:45 (UTC)

Not building for me, but filling tmpfs / memory until oom kills it.